发表机构
Shanghai Jiao Tong University; Alibaba Group; Tianji KernalMind Co., Ltd.; The Hong Kong University of Science and Technology; Southeast University; Renmin University of China; The University of Tokyo（上海交通大学; 阿里巴巴集团; 天机芯智有限公司; 香港科技大学; 东南大学; 中国人民大学; 东京大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出EgoGenesis模拟器，通过在线锚定投影记忆与Action-3D RoPE合成自我中心操作视频，扩充训练数据，显著提升了真实机器人单臂、双臂任务的分布外操作成功率。
AI 中文摘要
自我中心视频为具身人工智能提供了丰富的操作体验，但跨场景、物体、动作和具身体收集多样化的自我中心数据成本高昂。本文提出EgoGenesis，一种自我中心世界-动作模拟器，可合成可控、高质量的操作视频以扩充稀缺的真实世界训练数据。该方法基于预训练视频生成先验，引入两种几何感知条件机制：在线锚定投影记忆（OAPM）在自回归生成时保留首帧3D场景锚，同时定期刷新近期状态；Action-3D旋转位置嵌入（A3D-RoPE）用相机感知的3D旋转坐标编码末端执行器运动，将动作几何注入骨架到视频的交叉注意力以实现精确控制。这些组件共同提升了长序列自我中心生成的视觉保真度、几何稳定性和动作对齐度。此外，用400条EgoGenesis生成的轨迹增强400条真实轨迹后，单臂任务的分布外真实机器人成功率从77%提升至84%，双臂任务从53%提升至70%，证明合成数据可显著提升下游世界-动作建模（WAM）的泛化能力。
英文摘要
Egocentric video offers rich manipulation experience for embodied AI, yet collecting diverse egocentric data across scenes, objects, motions, and embodiments remains costly. We present \method, an egocentric world-action simulator that synthesizes controllable, high-quality manipulation videos to expand scarce real-world training data. \method{} builds on a pretrained video generation prior and introduces two geometry-aware conditioning mechanisms. Online Anchored Projective Memory (OAPM) preserves a first-frame 3D scene anchor while periodically refreshing a recent state during autoregressive generation. Action-3D Rotary Position Embedding (A3D-RoPE) encodes end-effector motion with camera-aware 3D rotary coordinates, injecting action geometry into skeleton-to-video cross-attention for precise control. Together, these components improve visual fidelity, geometric stability, and action alignment in long egocentric rollouts. Moreover, augmenting 400 real trajectories with 400 \method-generated trajectories improves out-of-distribution real-robot success from 77\% to 84\% on single-arm tasks and from 53\% to 70\% on dual-arm tasks, demonstrating that the synthesized data substantially improve downstream WAM generalization.
Commentsproject page: https://egogenesis.github.io/
